CN109886746A - A kind of trip purpose recognition methods based on passenger getting off car when and where - Google Patents
A kind of trip purpose recognition methods based on passenger getting off car when and where Download PDFInfo
- Publication number
- CN109886746A CN109886746A CN201910125740.0A CN201910125740A CN109886746A CN 109886746 A CN109886746 A CN 109886746A CN 201910125740 A CN201910125740 A CN 201910125740A CN 109886746 A CN109886746 A CN 109886746A
- Authority
- CN
- China
- Prior art keywords
- interest
- point
- trip
- passenger
- trip purpose
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of trip purpose recognition methods based on passenger getting off car when and where, comprising the following steps: (1) trip purpose is divided into several classifications, and be associated with related interests point；(2) whole day is divided into several periods, the ratio that trip purpose Zhan each in day part is always gone on a journey is as the time weighting of corresponding point of interest；(3) obtain usual maximum walking distance of the passenger after getting off, calculate effective point of interest to out-of-the-car position manhatton distance, in conjunction with distance and point of interest itself the weight calculation point of interest to the sucting strength of passenger；(4) classification summation is carried out to the sucting strength of point of interest relevant to all kinds of trip purposes, is the actual purpose of this trip by the maximum trip purpose kind judging of sucting strength.The present invention overcomes the defects that existing method only considers space attribute, improve the accuracy of trip purpose identification, cannot be only used for urban planning field, it can also be used to intelligent recommendation system.
Description
Technical field
The present invention relates to a kind of trip purpose recognition methods, belongs to field of intelligent transportation technology more particularly to one kind is based on
The trip purpose recognition methods of passenger getting off car when and where.
Background technique
In recent years, net about vehicle gradually becomes a kind of important trip mode in people's daily life, built-in vehicle
Positioning device has recorded a large amount of location information, but we have no way of learning to the trip purpose of passenger, however trip purpose is known
Not such as urban planning, shopping recommend and in terms of be of great significance.
Point of interest is the point for describing specific activities and service location, and geographic object a little can be abstracted as by referring to all,
More especially live closely related geographical entity, such as school, bank, restaurant, gas station, hospital, supermarket with people.It is emerging
Interest point largely represents the situation of land use in a region, therefore is often used to carry out the knowledge of trip purpose
Not.Existing many researchs combine the time attribute and space attribute of passenger's trip, but in the process of trip purpose identification
In still have many problems.On the one hand, space attribute is usually characterized with land use pattern, however a region use it is same
The land use pattern data of sample, so that the accurate expression of space attribute is to be improved.On the other hand, time attribute and space belong to
The combination degree of property is to be improved.
Summary of the invention
Goal of the invention: the present invention proposes a kind of method for identifying trip purpose, and when getting off of passenger is utilized in time
Between data characterize time attribute, spatially used the point of interest near out-of-the-car position to characterize space attribute, will be different
Trip purpose classification is associated with point of interest, calculates different trip purpose proportions in day part, in conjunction with travel activity when
Between attribute and space attribute, accurately calculate this trip and belong to the probability of all kinds of purposes, and then identify the trip mesh of this trip
's.
Technical solution: a kind of trip purpose recognition methods based on passenger getting off car when and where of the present invention, packet
Include step:
(1) trip purpose of passenger is divided into several classifications, trip purpose and related interests point is associated.
Further, the step (1) includes:
(1.1) all points of interest within the scope of urban area are collected；
(1.2) classify according to trip purpose to point of interest.
(2) whole day is divided into some time at certain intervals, calculates in each period various trip mesh total
Shared ratio in trip, using this accounting as the time weighting of point of interest corresponding with such trip purpose.
Further, the step (2) includes:
(2.1) some time was divided by 24 hours one day；
(2.2) obtain and ride on and off time period t belonging to the moment, t=1,2 ..., 24 and t-th of period in corresponding m
The weight k of the point of interest of class trip purposetm, 0 < ktm<1；Wherein, the period is that the passenger getting off car moment is corresponding next
The hourage at a integral point moment；ktmThe ratio always gone on a journey for the trip Zhan of m classification corresponding in the t period.
(3) usual maximum walking distance of the passenger after getting off is obtained, calculates effective point of interest to the graceful of out-of-the-car position
Hatton's distance, in conjunction with distance and point of interest itself the weight calculation point of interest to the sucting strength of passenger；Wherein, described effectively emerging
Interest point is away from all point of interest within the maximum walking distance of out-of-the-car position.
Further, the step (3) is that a certain point of interest in maximum walking distance is calculated according to following formula to multiplying
The sucting strength of visitor:
Wherein, r is the maximum walking distance after passenger getting off car, and N is the quantity of all points of interest in distance r, and n is distance
The quantity of point of interest relevant to certain one kind trip purpose, d in riManhattan for out-of-the-car position away from point of interest a certain in r away from
From ktmFor the weight of each point of interest.
(4) classification summation is carried out to the sucting strength of point of interest relevant to all kinds of trip purposes, by sucting strength maximum
Trip purpose kind judging be this trip actual purpose.
Further, the step (4) includes:
(4.1) certain class trip purpose is calculated to the sucting strength of passenger according to following formula:
It (4.2) is the actual purpose of this trip of passenger by the maximum trip purpose kind judging of sucting strength.
The utility model has the advantages that the invention has the following advantages that
1, point of interest is associated with trip purpose, by the quantity for obtaining various points of interest in the walking distance of out-of-the-car position
And point of interest can finely characterize the space attribute of passenger getting off car position to the manhatton distance of out-of-the-car position；
2, according to the trip purpose of people, different periods generate the objective circumstances of variation in one day, obtain getting off for passenger
Time obtains accounting of the various trip purposes in total trip in day part, can finely characterize the time attribute of this trip；
3, with above-mentioned accounting to each point of interest assign weight, regard as the time attribute of each point of interest, the comprehensive consideration time
Attribute and space attribute can simply accurately identify the trip purpose of passenger.
Detailed description of the invention
Fig. 1 is overview flow chart of the invention.
Specific embodiment
Further description of the technical solution of the present invention with reference to the accompanying drawings and examples.
As shown in Figure 1, a kind of trip purpose recognition methods based on passenger getting off car when and where of the present invention: first
Trip purpose is first divided into several classifications, various trip purposes and related interests point are associated.Secondly with it is certain when
Between interval by one day be divided into several periods, obtain and calculate various trip purpose proportions in day part.Then under retrieving
All points of interest in the walking distance greatest around of truck position assign point of interest based on this period all kinds of trip purpose accountings and weighing
Weight, according to each point of interest away from the space length of out-of-the-car position and each point of interest of its own weight calculation to the sucting strength of passenger.
Classification summation finally is carried out to the sucting strength of point of interest relevant to all kinds of trip purposes, by the maximum trip mesh of sucting strength
Kind judging be this trip actual purpose.
1) trip purpose and related interests point are associated with:
In conjunction with existing research achievement, the trip purpose of passenger is divided into six big classifications: going home, work, educate, purchase
Object, leisure and other.Every a kind of trip purpose and the progress of relevant point of interest is classification associated, as shown in table 1.
1 trip purpose of table and its related interests point
| Trip purpose | Related interests point |
| It goes home | Cell, apartment, dormitory etc. |
| Work | Company, factory etc. |
| Education | Universities, kindergarten etc. |
| Shopping | Supermarket, shop etc. |
| Leisure | Museum, cinema, park etc. |
| Other | Means of transportation, government organs etc. |
2) different trip purpose accountings and point of interest weighting in day part:
With 1 hour for interval, whole day is divided into 24 periods.It obtains and calculates different trip purposes in day part and account for
The ratio k always to go on a journeytm, wherein t represents passenger and completes to get off the period belonging to the moment when this goes on a journey by bus, and m represents trip
Purpose classification has all tT is integer as the period, and an entire period corresponds to identical
Number.Such as 0:01-0:59 corresponds to the period 1, i.e., t=1 at this time.
All kinds of trip purpose accountings are as shown in table 2 in day part.
2 Time segments division of table and each trip purpose accounting
Each point of interest one power is assigned according to the corresponding relationship of point of interest and trip purpose in table 1 in different periods
Weight.ktmThe weight of the point of interest of corresponding m class trip purpose as in the t period.The point of interest for imparting weight is used for next
Single point of interest is calculated in step to the sucting strength of passenger.
3) in walking distance single point of interest sucting strength:
The passenger to go on a journey by bus generally will not set a maximum walking distance r too far in later walking distance of getting off,
Represent the maximum walking distance of passenger getting off car later under normal circumstances.diIndicate Man Ha of the out-of-the-car position away from point of interest a certain in r
Distance.N indicates the quantity of all points of interest in distance r, and n indicates the interior point of interest relevant to certain one kind trip purpose of this distance
Quantity.Generally, passenger getting off car position can be near actual purpose ground, and closer point of interest is passenger's sheet with out-of-the-car position
Secondary trip purpose a possibility that it is bigger.Only consider influence of the out-of-the-car position space attribute to trip purpose, indicates a certain emerging with V
Interest point to the sucting strength of passenger, in other words a certain point of interest be the actual purpose of this trip a possibility that size, then have:
Various trip purpose proportions are different in different periods, in other words for traveler, same interest
Point is different in sucting strength in different time periods.Indicate that this combines the sucting strength of time attribute with V ', then at this time
A certain point of interest to the sucting strength of the passenger that gets off then are as follows:
4) the trip purpose identification of comprehensive all points of interest:
It sums to the sucting strength of point of interest relevant to certain class trip purpose, as such trip purpose multiplies this
The sucting strength of visitor:
It is the actual purpose of this trip by the maximum trip purpose kind judging of sucting strength.
Embodiment
Example is carried out with the net in Nanjing about vehicle trip data.It is related to the total of in August, 2018 in September, -2018 altogether
159 trips record, there is the detailed beginning and ending time by bus in order of going on a journey, out-of-the-car position and this time ride trip purpose by
Passenger provides.The aforementioned various trip purposes that the log data set is related to, detailed amount are as shown in table 3.
The quantity on order of the different trip purposes of table 3
| Trip purpose | It goes home | Work | Education | Shopping | Leisure | Other |
| Quantity on order | 48 | 39 | 8 | 32 | 15 | 17 |
(1) model foundation
It obtains and calculates the different trip purpose accounting k in each periodtmAs shown in table 4.
The ratio of different trip purposes in 4 day part of table
In conjunction with existing disclosed research achievement, the maximum walking distance r after passenger getting off car is set as 200 meters, i.e., for
Go on a journey each time, it is believed that walking distance of the passenger after getting off away from its destination be no more than 200 meters, so only retrieve away from
Point of interest of the out-of-the-car position manhatton distance no more than 200 meters is effective point of interest.With the data in table 4 to all effective interest
The V of point is weighted to obtain V '.Sucting strength by calculating all kinds of effective points of interest determines trip purpose.
(2) effect analysis
Trip purpose recognition result is as shown in table 5.Among 159 datas of collection, 135 trip purpose identification is obtained
Correct result, overall accuracy have reached 84.9%.Compared to the identification method (accuracy rate only with interest point data
67%), passenger's trip purpose of binding time attribute and space attribute identification achieves more good effect.
5 trip purpose recognition result of table
Claims (5)
1. a kind of trip purpose recognition methods based on passenger getting off car when and where, which is characterized in that comprising steps of
(1) trip purpose of all trip passengers by bus is divided into several classifications, trip purpose and related interests point is carried out
Association；
(2) whole day is divided into some time at certain intervals, calculates various trip mesh in each period and goes on a journey always
In shared ratio, using this accounting as the time weighting of point of interest corresponding with such trip purpose；
(3) statistics obtains usual maximum walking distance of the passenger after getting off, and calculates effective point of interest to the graceful of out-of-the-car position
Hatton's distance, in conjunction with distance and point of interest itself the weight calculation point of interest to the sucting strength of passenger；Wherein, described effectively emerging
Interest point is away from all point of interest within the maximum walking distance of out-of-the-car position；
(4) classification summation is carried out to the sucting strength of point of interest relevant to all kinds of trip purposes, sucting strength is maximum out
Row purpose kind judging is the actual purpose of this trip.
2. the trip purpose recognition methods according to claim 1 based on passenger getting off car when and where, which is characterized in that
The step (1) includes:
(1.1) all points of interest within the scope of urban area are collected；
(1.2) classify according to trip purpose to point of interest.
3. the trip purpose recognition methods according to claim 1 based on passenger getting off car when and where, which is characterized in that
The step (2) includes:
(2.1) 24 periods were divided by 24 hours one day；
(2.2) obtain and ride on and off time period t belonging to the moment, t=1,2 ..., 24 and t-th of period in corresponding m class go out
The weight k of the point of interest of row purposetm, 0 < ktm<1；Wherein, the period is that the passenger getting off car moment is corresponding next whole
The hourage of point moment；ktmThe ratio always gone on a journey for the trip Zhan of m classification corresponding in the t period.
4. the trip purpose recognition methods according to claim 1 based on passenger getting off car when and where, which is characterized in that
The step (3) is that a certain point of interest in maximum walking distance is calculated according to following formula to the sucting strength of passenger:
Wherein, r is the maximum walking distance after passenger getting off car, and N is the quantity of all points of interest in distance r, and n is in distance r
The quantity of point of interest relevant to certain one kind trip purpose, diManhatton distance for out-of-the-car position away from point of interest a certain in r, ktm
For the weight of each point of interest.
5. the trip purpose recognition methods according to claim 1 based on passenger getting off car when and where, which is characterized in that
The step (4) includes:
(4.1) certain class trip purpose is calculated to the sucting strength of passenger according to following formula:
It (4.2) is the actual purpose of this trip of passenger by the maximum trip purpose kind judging of sucting strength.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910125740.0A CN109886746B (en) | 2019-02-20 | 2019-02-20 | Trip purpose identification method based on time and place of passenger getting off |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910125740.0A CN109886746B (en) | 2019-02-20 | 2019-02-20 | Trip purpose identification method based on time and place of passenger getting off |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109886746A true CN109886746A (en) | 2019-06-14 |
| CN109886746B CN109886746B (en) | 2023-04-25 |
Family
ID=66928564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910125740.0A Active CN109886746B (en) | 2019-02-20 | 2019-02-20 | Trip purpose identification method based on time and place of passenger getting off |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886746B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832768A (en) * | 2019-08-13 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | POI feature generation method and device, electronic equipment and storage medium |
| CN113642625A (en) * | 2021-08-06 | 2021-11-12 | 北京交通大学 | Method and system for deducing individual trip purpose of urban rail transit passenger |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103413178A (en) * | 2013-07-13 | 2013-11-27 | 北京工业大学 | Sampling-based park-and-ride facility attraction demand quantitative classification calculation method |
| CN107749020A (en) * | 2017-10-23 | 2018-03-02 | 重庆大学 | A kind of commending system based on supposition cab-getter's trip purpose |
-
2019
- 2019-02-20 CN CN201910125740.0A patent/CN109886746B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103413178A (en) * | 2013-07-13 | 2013-11-27 | 北京工业大学 | Sampling-based park-and-ride facility attraction demand quantitative classification calculation method |
| CN107749020A (en) * | 2017-10-23 | 2018-03-02 | 重庆大学 | A kind of commending system based on supposition cab-getter's trip purpose |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832768A (en) * | 2019-08-13 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | POI feature generation method and device, electronic equipment and storage medium |
| CN113642625A (en) * | 2021-08-06 | 2021-11-12 | 北京交通大学 | Method and system for deducing individual trip purpose of urban rail transit passenger |
| CN113642625B (en) * | 2021-08-06 | 2024-02-02 | 北京交通大学 | Method and system for deducing individual travel purposes of urban rail transit passengers |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109886746B (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhong et al. | Inferring building functions from a probabilistic model using public transportation data | |
| CN110533038A (en) | A method of urban vitality area and inner city Boundary Recognition based on information data | |
| Prillwitz et al. | Moving towards sustainability? Mobility styles, attitudes and individual travel behaviour | |
| CN108133302B (en) | Public bicycle potential demand prediction method based on big data | |
| Kim et al. | Measuring the benefits of air quality improvement: a spatial hedonic approach | |
| Guo et al. | Modifiable areal units: Problem or perception in modeling of residential location choice? | |
| CN112133090A (en) | Multi-mode traffic distribution model construction method based on mobile phone signaling data | |
| Berechman et al. | Evaluation, prioritization and selection of transportation investment projects in New York City | |
| CN107729938B (en) | Rail station classification method based on bus connection radiation zone characteristics | |
| Bhat et al. | Accommodating immigration status and self-selection effects in a joint model of household auto ownership and residential location choice | |
| CN110414751A (en) | A kind of hotel industry addressing evaluation system and evaluation method based on geographical location | |
| Zhu et al. | Understanding the spatiotemporal patterns of public bicycle usage: A case study of Hangzhou, China | |
| CN108717676A (en) | Evaluation space method and system are lived in duty under different scale based on multi-data fusion | |
| CN112785029B (en) | Charging station power consumption prediction method based on deep clustering model | |
| Nam et al. | Analysis of grid cell–based taxi ridership with large-scale GPS data | |
| Montini et al. | Route and mode choice models using GPS data | |
| CN109886746A (en) | A kind of trip purpose recognition methods based on passenger getting off car when and where | |
| Zhang et al. | Detecting taxi travel patterns using GPS trajectory data: A case study of Beijing | |
| Held et al. | Future mobility demand estimation based on sociodemographic information: A data-driven approach using machine learning algorithms | |
| Vrtic et al. | The impact of tilting trains in Switzerland: A route choice model of regional-and long distance public transport trips | |
| CN111563666A (en) | Urban public service facility space configuration evaluation method based on network heat | |
| CN108182196A (en) | A kind of Urban traffic demand Forecasting Methodology based on POI | |
| CN116502837B (en) | Sports facility configuration method and device based on big data and storage medium | |
| Chung et al. | Investigating the effects of POI-based land use on traffic accidents in Suzhou Industrial Park, China | |
| Zhou et al. | Big data for intrametropolitan human movement studies A case study of bus commuters based on smart card data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |